The study will identify patterns of illicit psychoactive drug use in suburban children linking age of onset, type of use and continuity of use with school performance, child background, and family characteristics. The impact of three kinds of drug education on drug use onset and continuation will be assessed, as will other changes over a two year follow-up. The sample will comprise about 2100 elementary school and 1200 high school students taken from grades 2, 4,6, 8 and 10. They will be randomly assigned to three types of drug education in school; minimal, didactic and process. Results will provide data on what kind of child begins what kind of drug use when, what happens over two years, and how education affects that use.